Re: [PATCH net v1 1/1] net: dsa: microchip: KSZ9893: do not write to not supported Output Clock Control Register

[Andrew Lunn <andrew@lunn.ch>]

On Thu, Jul 28, 2022 at 03:18:52PM +0200, Oleksij Rempel wrote:
> KSZ9893 compatible chips do not have "Output Clock Control Register 0x0103".
> So, avoid writing to it.
> 
> Fixes: 462d525018f0 ("net: dsa: microchip: move ksz_chip_data to ksz_common")
> Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
> ---
>  drivers/net/dsa/microchip/ksz9477.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/net/dsa/microchip/ksz9477.c b/drivers/net/dsa/microchip/ksz9477.c
> index 5dff6c3279bb..c73bb6d383ad 100644
> --- a/drivers/net/dsa/microchip/ksz9477.c
> +++ b/drivers/net/dsa/microchip/ksz9477.c
> @@ -198,6 +198,10 @@ int ksz9477_reset_switch(struct ksz_device *dev)
>  	ksz_write32(dev, REG_SW_PORT_INT_MASK__4, 0x7F);
>  	ksz_read32(dev, REG_SW_PORT_INT_STATUS__4, &data32);
>  
> +	/* KSZ9893 compatible chips do not support refclk configuration */
> +	if (dev->chip_id == KSZ9893_CHIP_ID)
> +		return 0;
> +

Do you actually want to return -EINVAL? I assume this is being driven
by a DT property? And that property is not valid for this chip. So we
want to let the DT writer know. Question is, is there a backwards
compatibility issue? If this has always been silently ignored, and
there are DT with this property, do we want to break them.

      Andrew